Unsupervised Learning Capstone

The aim of this capstone is to cluster our data set with unsupervised clustering methods and alternative models. For this purpose we used a data from a web site Turo.com, the biggest car sharing platform in the USA. There are two problems that required unsupervised learning methods:

1. We need to cluster our data by location
2. Categorize all cars by their characteristics. This information will help to continue the supervised learning and to create the predictive models.

Location clustering For the purpose of location clustering we used the Latitude and Longtitude assosiated with every listing in the data. We approached our dataset with k-means, dbscan and hierarhical clustering techniques. We will use the siloughette score for each of the methodics to define how precise is every method.

Car category clustering For this purpose we manually assigned categories for each car. We will try to do clustering with unsupervised clustering techniques ( K-means, Hierarchical Clustering and DBSCAN) with dimensionality reduction such as PSA and UMAP but more likely the best way will be to procees with the supervised classifier model. We will use neural network multi layer perceptron model and random forest to see what performs better.

#here is how our data looking
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 168109 entries, 0 to 168108
Data columns (total 100 columns):
 #   Column                               Non-Null Count   Dtype  
---  ------                               --------------   -----  
 0   Car ID                               168109 non-null  int64  
 1   URL                                  168109 non-null  object 
 2   Make                                 168109 non-null  object 
 3   Model                                168108 non-null  object 
 4   Trim                                 44008 non-null   object 
 5   Year                                 168109 non-null  int64  
 6   Color                                60377 non-null   object 
 7   Transmission                         168109 non-null  object 
 8   Fuel Type                            70907 non-null   object 
 9   Number of seats                      69655 non-null   float64
 10  Number of doors                      70911 non-null   float64
 11  GPS                                  61296 non-null   float64
 12  Convertible                          10764 non-null   float64
 13  Status                               168109 non-null  object 
 14  Booking Instantly                    168109 non-null  int64  
 15  City                                 168109 non-null  object 
 16  State                                168109 non-null  object 
 17  ZIP Code                             167789 non-null  object 
 18  Country                              168109 non-null  object 
 19  Latitude                             168109 non-null  float64
 20  Longitude                            168109 non-null  float64
 21  Owner ID                             168109 non-null  int64  
 22  Owner                                168109 non-null  object 
 23  Price per day                        168109 non-null  float64
 24  Custom Delivery fee                  52994 non-null   object 
 25  Airport Delivery fee                 15637 non-null   object 
 26  Distance Included/day, Miles, KM     163241 non-null  float64
 27  Distance Included/Week, Miles, KM    163237 non-null  float64
 28  Distance Included/Monthy, Miles, KM  162457 non-null  float64
 29  Booking Discount - Weekly            168109 non-null  int64  
 30  Booking Discount - Monthly           168109 non-null  int64  
 31  Fee for extra mile, $                168109 non-null  float64
 32  Registration date                    168109 non-null  object 
 33  Trip Count                           168109 non-null  int64  
 34  Reviews number                       168109 non-null  int64  
 35  Owner rate                           57169 non-null   float64
 36  Airport ride needed                  14615 non-null   object 
 37  Business class                       168109 non-null  bool   
 38  Vehicle protection                   168109 non-null  object 
 39  numberOfFaqs                         168109 non-null  int64  
 40  regularAirportDeliveryFee            15637 non-null   object 
 41  minimumAgeInYearsToRent              5332 non-null    float64
 42  numberOfFavorites                    168109 non-null  int64  
 43  highValueVehicle                     168109 non-null  bool   
 44  frequentlyBooked                     168109 non-null  bool   
 45  dateRangeRate                        0 non-null       float64
 46  Occupancy Jan                        21519 non-null   float64
 47  Occupancy Feb                        29792 non-null   float64
 48  Occupancy Mar                        23022 non-null   float64
 49  Occupancy Apr                        13271 non-null   float64
 50  Occupancy May                        14062 non-null   float64
 51  Occupancy Jun                        15558 non-null   float64
 52  Occupancy Jul                        18308 non-null   float64
 53  Occupancy Aug                        18714 non-null   float64
 54  Occupancy Sep                        18513 non-null   float64
 55  Occupancy Oct                        18984 non-null   float64
 56  Occupancy Nov                        19799 non-null   float64
 57  Occupancy Dec                        23754 non-null   float64
 58  Another Occupancy Jan                15154 non-null   float64
 59  Another Occupancy Feb                25544 non-null   float64
 60  Another Occupancy Mar                24756 non-null   float64
 61  Another Occupancy Apr                9164 non-null    float64
 62  Another Occupancy May                9899 non-null    float64
 63  Another Occupancy Jun                11056 non-null   float64
 64  Another Occupancy Jul                12730 non-null   float64
 65  Another Occupancy Aug                13072 non-null   float64
 66  Another Occupancy Sep                12402 non-null   float64
 67  Another Occupancy Oct                12790 non-null   float64
 68  Another Occupancy Nov                13124 non-null   float64
 69  Another Occupancy Dec                15754 non-null   float64
 70  Unnamed: 70                          0 non-null       float64
 71  Unnamed: 71                          0 non-null       float64
 72  Unnamed: 72                          0 non-null       float64
 73  Unnamed: 73                          0 non-null       float64
 74  Unnamed: 74                          0 non-null       float64
 75  Unnamed: 75                          0 non-null       float64
 76  Unnamed: 76                          0 non-null       float64
 77  Unnamed: 77                          0 non-null       float64
 78  Unnamed: 78                          0 non-null       float64
 79  Unnamed: 79                          0 non-null       float64
 80  Unnamed: 80                          0 non-null       float64
 81  Unnamed: 81                          0 non-null       float64
 82  Unnamed: 82                          0 non-null       float64
 83  Unnamed: 83                          0 non-null       float64
 84  Unnamed: 84                          0 non-null       float64
 85  Unnamed: 85                          0 non-null       float64
 86  Unnamed: 86                          0 non-null       float64
 87  Unnamed: 87                          0 non-null       float64
 88  Unnamed: 88                          0 non-null       float64
 89  Unnamed: 89                          0 non-null       float64
 90  Unnamed: 90                          0 non-null       float64
 91  Unnamed: 91                          0 non-null       float64
 92  Unnamed: 92                          0 non-null       float64
 93  Unnamed: 93                          0 non-null       float64
 94  Unnamed: 94                          0 non-null       float64
 95  Unnamed: 95                          0 non-null       float64
 96  Unnamed: 96                          0 non-null       float64
 97  Unnamed: 97                          0 non-null       float64
 98  Unnamed: 98                          0 non-null       float64
 99  Unnamed: 99                          0 non-null       float64
dtypes: bool(3), float64(68), int64(10), object(19)
memory usage: 124.9+ MB

Basic data exploration

At this stage we will check the sanity of the data, make observations on the data type, missing values, shape of the data

# Remove all columns between column index 70 to 99 
df.drop(df.iloc[:, 69:100], inplace = True, axis = 1) 
#replace all the missing data with 0 (we'll be checking later if we need to adjust this 0 for some variables)
df = df.fillna(0)

#mapping our location distribution
import matplotlib.pyplot as plt
plt.scatter(x=df['Longitude'], y=df['Latitude'])
plt.show()

#drop all non-active car listings
df = df[df['Status'] == "Active"]

#Let's scale the variables 
scaler = StandardScaler()
df[['Latitude']] = scaler.fit_transform(df[['Latitude']])
df[['Longitude']] = scaler.fit_transform(df[['Longitude']])

For car clusterization purposes we're going to use these variables:

• 'Price per day'
• 'Year'
• 'Number of seats'
• 'Number of doors'
• 'Convertible'
• 'Fuel Type'
• 'Make'
• 'highValueVehicle'
• 'Transmission'
• "MinAge"

Our target pre-labeled variable is

• "Category"

Introducing Category variable

#combining make and model into category name
df1 = pd.read_csv ('Category_DB.csv', encoding='ISO-8859–1')
df['for category']= df['Make']+df['Model']
df = pd.merge (df, df1[['Car ID', 'Category']], on  = 'Car ID', how="left")

Clean the variables if needed

#adjust some text errors
df.Make = df.Make.str.lower()
df.Model = df.Model.str.lower()
df.Category = df.Category.astype('category')
df.Category = df.Category.str.lower()
df = df[df['Category'].notnull()]

df.Category = df.Category.astype('category')
df['Category_code'] = df['Category'].cat.codes

df.groupby("Category")["Category"].count()

Category
cargo van          25
convertible      2358
economy          6518
electric         1521
exotic           3811
intermediate    11127
minivan          1681
motorhome           5
premium          4804
premium suv      4427
retro             437
suv              5498
truck            1363
van               103
Name: Category, dtype: int64

#Here is the categorized distribution of the variable:
sns.countplot(x=df['Category_code']);

#Here is the categorized distribution of the variable:
sns.countplot(x=df['Number of seats']);

sns.countplot(x=df['Fuel Type']);

sns.countplot(x=df['Number of doors']);

sns.countplot(x=df['Convertible']);

sns.countplot(x=df['minimumAgeInYearsToRent']);

Let's do some work for null variables:

df["Number of seats"].fillna(df["Number of seats"].mean(), inplace=True)
df["Number of doors"].fillna(df["Number of doors"].mean(), inplace=True)
df['Fuel Type'] = df['Fuel Type'].astype(str)
df['Fuel Type'] = df['Fuel Type'].apply( lambda x: x.replace ('0', 'Gas'))

Potential outliers

#Checking pricing by categories
sns.set(style="ticks", palette="pastel")


# Draw a nested boxplot to show bills by day and time
chart = sns.boxplot(x="Category_code", y="Price per day", palette=["m", "g"],
            data=df)

We decided not to remove outliers because of the nature of data and actual important output from the car owners about the price if very significant and can be truthful.

CLUSTERIZATION

k-means clustering for location

The first problem that we're solving is to be able to separate our listings by location. We would like to use k-means because our data has only two variables that are measurements of distance. K-means should be the best fir for it however we're going to validate with other methods as well.

# Variable with the Longitude and Latitude
Z = df[['Car ID','Latitude','Longitude']]
K_clusters = range(1,20)
kmeans = [KMeans(n_clusters=i) for i in K_clusters]
Y_axis = df[['Latitude']]
X_axis = df[['Longitude']]
score = [kmeans[i].fit(Y_axis).score(Y_axis) for i in range(len(kmeans))]
# Visualize the amount of clusters
plt.plot(K_clusters, score)
plt.xlabel('Number of Clusters')
plt.ylabel('Score')
plt.title('Elbow Curve')
plt.show()

#Let's define labels for location
kmeans = KMeans(n_clusters = 12, init ='k-means++')
kmeans.fit(Z[Z.columns[1:3]]) # Compute k-means clustering.
Z['cluster_label'] = kmeans.fit_predict(Z[Z.columns[1:3]])
centers = kmeans.cluster_centers_ # Coordinates of cluster centers.
labels = kmeans.predict(Z[Z.columns[1:3]]) # Labels of each point

#put them on the plot
Z.plot.scatter(x = 'Latitude', y = 'Longitude', c=labels, s=50, cmap='viridis')
plt.scatter(centers[:, 0], centers[:, 1], c='black', s=200, alpha=0.5)

<matplotlib.collections.PathCollection at 0x1a2dbff0d0>

#This is our silhouette score for k-means
print(metrics.silhouette_score(Z[Z.columns[1:3]], labels, metric='euclidean'))

0.6456396547584401

DBSCAN Clustering for Location

# Defining the agglomerative clustering
dbscan_cluster_l = DBSCAN(eps=1, min_samples=5)

# Fit model
clusters = dbscan_cluster_l.fit_predict(Z[Z.columns[1:3]])

print(metrics.silhouette_score(Z[Z.columns[1:3]], clusters, metric='euclidean'))

0.0023567980682241657

Z.plot.scatter(x = 'Latitude', y = 'Longitude', c = clusters, s=50, cmap='viridis')
plt.scatter(centers[:, 0], centers[:, 1], c='black', s=200, alpha=0.5)

<matplotlib.collections.PathCollection at 0x1a366de290>

This is definitely won't be our choise based on the silhougtette score

Hierarchical clustering for location

from sklearn.cluster import AgglomerativeClustering

# Defining the agglomerative clustering
agg_cluster = AgglomerativeClustering(linkage='complete', 
                                      affinity='cosine',
                                      n_clusters=5)
# Fit model
clusters_h = agg_cluster.fit_predict(Z[Z.columns[1:3]])

Z.plot.scatter(x = 'Latitude', y = 'Longitude', c = clusters_h, s=50, cmap='viridis')
plt.scatter(centers[:, 0], centers[:, 1], c='black', s=200, alpha=0.5)

<matplotlib.collections.PathCollection at 0x1a28b7dd10>

print(metrics.silhouette_score(Z[Z.columns[1:3]], clusters_h, metric='euclidean'))

0.43172051126894256

This result is better but still worth than K-means

Performance of location clustering algorithm

We found that k-means work the best way for the location clustering. The siloughette score is 0.65. Let's review the clusters:

### Adding location cluster into the dataframe and will do a visual inspetion
df = df.merge(Z, left_on='Car ID', right_on='Car ID')

sns.countplot(x=labels);

df_location = df[['Car ID', 'City', 'Latitude_y', 'Longitude_y', 'State', 'cluster_label_y']]

df_location.head(20)

	Car ID	City	Latitude_y	Longitude_y	State	cluster_label_y
0	92	San Francisco	0.240971	-0.820969	CA	8
1	279	Hillsborough	0.216854	-0.818469	CA	8
2	445	Berkeley	0.251648	-0.815284	CA	8
3	724	San Jose	0.188699	-0.803155	CA	8
4	972	Santa Ana	-0.258606	-0.673040	CA	3
5	1081	San Leandro	0.235544	-0.811849	CA	8
6	1085	Sunnyvale	0.192158	-0.807857	CA	8
7	1461	Berkeley	0.251581	-0.815273	CA	8
8	1745	Los Angeles	-0.206255	-0.695349	CA	3
9	1759	Evansville	0.266085	0.313165	IN	7
10	1869	Laguna Niguel	-0.290516	-0.666406	CA	3
11	1921	Marina del Rey	-0.231929	-0.691059	CA	3
12	2450	Springfield	0.512054	0.710972	PA	1
13	2635	Los Angeles	-0.216662	-0.687127	CA	3
14	3056	Ontario	-0.222611	-0.663496	CA	3
15	3394	Charlotte	-0.076370	0.531650	NC	7
16	4200	Oshkosh	1.024917	0.281778	WI	1
17	4845	West Covina	-0.221915	-0.673850	CA	3
18	5009	Pleasant Hill	0.265646	-0.808728	CA	8
19	5037	Austin	-0.683902	-0.016798	TX	11

#TOP 10 state and cities by quantity of cars listed 
df.groupby(['State', 'City']).size().sort_values(ascending=False).head(10)

State  City         
CA     Los Angeles      3833
FL     Miami            1141
CA     San Diego         849
TX     Houston           759
NV     Las Vegas         747
CA     San Francisco     718
CO     Denver            654
IL     Chicago           650
HI     Honolulu          644
GA     Atlanta           526
dtype: int64

The results looks reliable visually. We can continue to inspect them.

WORK ON CLASTERISATION AND DIMENSIONALITY REDUCTION

Let's try to solve the problem of car segmentations based on their parameters with unsupervised clustering techniques.

Trsnsform non-numerical variables into dummies

df = pd.concat([df,pd.get_dummies(df.Convertible, prefix="Conv", drop_first=True)], axis=1)
df = pd.concat([df,pd.get_dummies(df['Fuel Type'], prefix="fuel", drop_first=True)], axis=1)
df = pd.concat([df,pd.get_dummies(df['Make'], prefix="Make", drop_first=True)], axis=1)
df = pd.concat([df,pd.get_dummies(df['highValueVehicle'], prefix="value", drop_first=True)], axis=1)
df = pd.concat([df,pd.get_dummies(df['Transmission'], prefix="Trans", drop_first=True)], axis=1)

dummy_column_names = list(pd.get_dummies(df.Convertible, prefix="Conv", drop_first=True).columns)
dummy_column_names = dummy_column_names + list(pd.get_dummies(df['Fuel Type'], prefix="fuel", drop_first=True).columns)
dummy_column_names = dummy_column_names + list(pd.get_dummies(df['Make'], prefix="Make", drop_first=True).columns)
dummy_column_names = dummy_column_names + list(pd.get_dummies(df['highValueVehicle'], prefix="value", drop_first=True).columns)
dummy_column_names = dummy_column_names + list(pd.get_dummies(df['Transmission'], prefix="Trans", drop_first=True).columns)

Assign incoming and outcoming variables

X =  df[['Price per day', 'Year', 'Number of seats', 'Number of doors']+dummy_column_names]
#Let's keep one more set for the variable without price
X1 =  df[['Year', 'Number of seats', 'Number of doors']+dummy_column_names]
Y = df["Category_code"]

#Let's see how categories looking
Y.value_counts()/len(Y)

5     0.254751
2     0.149228
11    0.125876
8     0.109987
9     0.101355
4     0.087252
1     0.053986
6     0.038486
3     0.034823
12    0.031206
10    0.010005
13    0.002358
0     0.000572
7     0.000114
Name: Category_code, dtype: float64

# Reduce it to two components for visualization and check how long will it take 
X_pca = PCA(2).fit_transform(X_std)

time_start = time.time()
X_pca = PCA(2).fit_transform(X_std)


print('PSA done! Time elapsed: {} seconds'.format(time.time()-time_start))

PSA done! Time elapsed: 0.08752202987670898 seconds

UMAP

time_start = time.time()

umap_results = umap.UMAP(n_neighbors=5,
                      min_dist=0.3,
                      metric='correlation').fit_transform(X_std)

print('UMAP done! Time elapsed: {} seconds'.format(time.time()-time_start))

UMAP done! Time elapsed: 37.60747003555298 seconds

#we see that the UMAP technique is much longer and let's see how it's looking

plt.figure(figsize=(10,5))
plt.scatter(umap_results[:, 0], umap_results[:, 1])
plt.xticks([])
plt.yticks([])
plt.axis('off')
plt.show()

This isn't a good perfomace from a clustering perspective plus the calculation time is too high

umap_results7 = umap.UMAP(n_neighbors=7,
                      min_dist=0.3,
                      metric='correlation').fit_transform(X_std)

plt.figure(figsize=(10,5))
plt.scatter(umap_results7[:, 0], umap_results7[:, 1])
plt.xticks([])
plt.yticks([])
plt.axis('off')
plt.show()

6.1. K-Means

# Calculate predicted values.
y_pred = KMeans(n_clusters=9, random_state=123).fit_predict(X_std)

# Plot the solution.
plt.scatter(X_pca[:, 0], X_pca[:, 1],  c=y_pred)
plt.show()

K-means Evaluation

# Non-labeled verification:

labels = KMeans(n_clusters=7, random_state=123).fit_predict(X_std)
print(metrics.silhouette_score(X_pca, labels, metric='euclidean'))

0.13961588341366074

#the score is very low since our goal is to be closer to 1. Let's try another techniques

DBSCAN (psa)

from sklearn.cluster import DBSCAN
from sklearn.decomposition import PCA

scaler = StandardScaler()
X_std = scaler.fit_transform(X)
X_pca = PCA(2).fit_transform(X_std)
# Defining the agglomerative clustering
dbscan_cluster = DBSCAN(eps=1, min_samples=5)

# Fit model
clusters_db_pc = dbscan_cluster.fit_predict(X_std)

#plot the results
pca = PCA(n_components=2).fit_transform(X_std)

plt.figure(figsize=(10,5))
colours = 'rbg'

for i in range(pca.shape[0]):
    plt.text(pca[i, 0], pca[i, 1], str(clusters_db_pc[i]),
             fontdict={'weight': 'bold', 'size': 50}
        )
    
plt.xticks([])
plt.yticks([])
plt.axis('off')
plt.show()

print(metrics.silhouette_score(X_std, clusters_db_pc, metric='euclidean'))

0.5179374100527134

DBSCAN (umap)

#The score is much better. Let's try with UMAP dimensinality reduction

# Defining the agglomerative clustering
dbscan_cluster = DBSCAN(eps=1, min_samples=5)

# Fit model
clusters_umap = dbscan_cluster.fit_predict(umap_results7)

print(metrics.silhouette_score(X_std, clusters_umap, metric='euclidean'))

-0.48776061471306803

This isn't a good number so let's try another way

Hierarchical clastering (psa)

scaler = StandardScaler()
X_std = scaler.fit_transform(X)
Xpca = PCA(2).fit_transform(X_std)
from sklearn.cluster import AgglomerativeClustering

# Defining the agglomerative clustering
agg_cluster = AgglomerativeClustering(linkage='complete', 
                                      affinity='cosine',
                                      n_clusters=3)

# Fit model
clusters_h = agg_cluster.fit_predict(X_pca)

pca = PCA(n_components=2).fit_transform(X_std)

plt.figure(figsize=(10,5))
colours = 'rbg'
for i in range(pca.shape[0]):
    plt.text(pca[i, 0], pca[i, 1], str(clusters_h[i]),
             fontdict={'weight': 'bold', 'size': 50}
        )

plt.xticks([])
plt.yticks([])
plt.axis('off')
plt.show()

print(metrics.silhouette_score(X_std, clusters_h, metric='euclidean'))

0.07522371146298965

This one doesn't work this good, let's try with UMAP

Hierarchical clastering (umap)

print(metrics.silhouette_score(X_std, clusters_h_umap, metric='euclidean'))

-0.0018369254175829987

the rate is very low, we will be skipping it.

UNSUPERVISED CLUSTERING. CONCLUSION

During unsupervised clustering process we were unable to define clear unsupervised clustering solution for our problem. However, based on the silhouette score the best result is coming from DBSCAN clasterisation using pca technique was high we still can't rely on it. We need to go with more sophisticated approach since our variables contain mixed (numeric and non-numeric and categorical) types of data. Also, there are some sort of neural charachteristics for the clusters. In this case we need to move forward and try to create a classificator with neural network to be able to assign the category for all new cars that will be added into the system.

Multi Layer Perceptron model

The power of neural networks come from their ability to learn the representation in the training data and how to best relate it to the output variable that you want to predict. We will try to use MLP model as well as Random Forest Classifier and compare the performance

# Import the model.
from sklearn.neural_network import MLPClassifier

# Establish and fit the model, with a single, 1000 perceptron layer.
mlp = MLPClassifier(hidden_layer_sizes=(1000,))
mlp.fit(X, Y)

MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,
              beta_2=0.999, early_stopping=False, epsilon=1e-08,
              hidden_layer_sizes=(1000,), learning_rate='constant',
              learning_rate_init=0.001, max_iter=200, momentum=0.9,
              n_iter_no_change=10, nesterovs_momentum=True, power_t=0.5,
              random_state=None, shuffle=True, solver='adam', tol=0.0001,
              validation_fraction=0.1, verbose=False, warm_start=False)

mlp.score(X, Y)

0.6206099180365401

the score is pretty good. Let's try to do a cross validation for this

from sklearn.model_selection import cross_val_score
cross_val_score(mlp, X, Y, cv=5)

array([0.55227637, 0.56745623, 0.59512363, 0.50028634, 0.37651775])

The cross validation scores are around 0.54 in average.

Let's try it with the standartized X values - X_std

# Import the model.
from sklearn.neural_network import MLPClassifier

# Establish and fit the model, with a single, 1000 perceptron layer.
mlp = MLPClassifier(hidden_layer_sizes=(1000,))
mlp.fit(X_std, Y)

MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,
              beta_2=0.999, early_stopping=False, epsilon=1e-08,
              hidden_layer_sizes=(1000,), learning_rate='constant',
              learning_rate_init=0.001, max_iter=200, momentum=0.9,
              n_iter_no_change=10, nesterovs_momentum=True, power_t=0.5,
              random_state=None, shuffle=True, solver='adam', tol=0.0001,
              validation_fraction=0.1, verbose=False, warm_start=False)

mlp.score(X_std, Y)

0.7322679609872247

This is definitely much better. We've got our model explained by 10points better when our data is standartized

from sklearn.model_selection import cross_val_score
cross_val_score(mlp, X_std, Y, cv=5)

array([0.60123542, 0.62959149, 0.69013278, 0.72786622, 0.69873998])

This is a solid result for the data on average 0.69

Let's try another clasifier to see if our results can go higher

Random forest classifier

from sklearn import ensemble
from sklearn.model_selection import cross_val_score

rfc = ensemble.RandomForestClassifier()
cross_val_score(rfc, X, Y, cv=5)

array([0.56074125, 0.59903879, 0.65338828, 0.67678387, 0.65670103])

these numbers are good but the best performance we've got with the

CONCLUSIONS

In this capstone we tried to solve the problem of clustering the data. Our original raw data has variables that should be clustered in order to make more precise predictions and to reduce the amount of noise and overfitting. Unsupervised clustering is mainly a process of putting several objects into one bin while models is an ability to predict an outcome. Two issues were solved:

• clustering listings by location
• clustering by car characteristics (price, make, type of fuel, high/not high value of the car

1. Location clustering: The unsupervised learning worked well for clustering numerical information with Lat and Long (we got the silhouette score on the level of 0.65 with k-means .

2. Car category clustering we tried several unsupervised learning techniues but non of them, work good for hybrid variables. We were able to achieve pretty high silghouette score with dbscan (0.84) but it doesn't work good along with the labeled data. We tried another methods as well. K-means and hierarchical clustering are good at finding circular (or convex) clusters, which makes them great tools for identifying well-separated clusters. But, unfortunately, they're not good at identifying clusters that are not well-separated or that have non-convex shapes such as rings inside rings. Anyway, this problem better should be solved with the supervised model since there are more mixed data involved, it has non-linear connections. For this purpose we used supervised neural network classifier that gave us around 0.70 score. We check models with cross validation score to make sure there is no overfitting in the data These findings are extremelly useful for the development an investment tool in order to predict potential earnings from sharing cars on the platform. We're going to continue our work on this data set in order to increase the relaibality of our calculations.